Skip to content

dnn(test): adjust EAST test tolerance#19074

Merged
opencv-pushbot merged 1 commit intoopencv:masterfrom
alalek:dnn_test_tolerance_east
Dec 10, 2020
Merged

dnn(test): adjust EAST test tolerance#19074
opencv-pushbot merged 1 commit intoopencv:masterfrom
alalek:dnn_test_tolerance_east

Conversation

@alalek
Copy link
Copy Markdown
Member

@alalek alalek commented Dec 10, 2020

relates #17570

Error messages:

MYRIAD
[ RUN      ] Test_Model.TextDetectionByEAST/1, where GetParam() = NGRAPH/MYRIAD
/build/precommit_custom_linux/opencv/modules/dnn/test/test_model.cpp:258: Failure
The difference between box.center.x and gtBox.center.x is 4.37310791015625, which exceeds EPS_PIXELS, where
box.center.x evaluates to 661.923095703125,
gtBox.center.x evaluates to 657.54998779296875, and
EPS_PIXELS evaluates to 3.
/build/precommit_custom_linux/opencv/modules/dnn/test/test_model.cpp:260: Failure
The difference between box.size.width and gtBox.size.width is 18.09197998046875, which exceeds EPS_PIXELS, where
box.size.width evaluates to 298.74801635742188,
gtBox.size.width evaluates to 316.83999633789062, and
EPS_PIXELS evaluates to 3.
[  FAILED  ] Test_Model.TextDetectionByEAST/1, where GetParam() = NGRAPH/MYRIAD (1136 ms)
OpenCL FP16 (Intel i5-6600 iGPU)
[ RUN      ] Test_Model.TextDetectionByEAST/1, where GetParam() = OCV/OCL_FP16
/home/alalek/projects/opencv/dev/modules/dnn/test/test_model.cpp:260: Failure
The difference between box.size.width and gtBox.size.width is 9.49346923828125, which exceeds EPS_PIXELS, where
box.size.width evaluates to 307.34652709960938,
gtBox.size.width evaluates to 316.83999633789062, and
EPS_PIXELS evaluates to 3.
[ INFO:0] global /home/alalek/projects/opencv/dev/modules/ts/src/ts.cpp (848) testTearDown Memory_usage (OpenCL): 259735932 (base=0  current=4096)
[  FAILED  ] Test_Model.TextDetectionByEAST/1, where GetParam() = OCV/OCL_FP16 (2388 ms)
CUDA FP16
[ RUN      ] Test_Model.TextDetectionByEAST/1, where GetParam() = CUDA/CUDA_FP16
[ INFO:0] global C:\projects\opencv\dev\modules\dnn\src\dnn.cpp (2421) cv::dnn::dnn4_v20201117::Net::Impl::initCUDABackend CUDA backend will fallback to the CPU implementation for the layer "_input" of type __NetInputLayer__

C:\projects\opencv\dev\modules\dnn\test\test_model.cpp(258): error: The difference between box.center.x and gtBox.center.x is 4.224609375, which exceeds EPS_PIXELS, where
box.center.x evaluates to 661.77459716796875,
gtBox.center.x evaluates to 657.54998779296875, and
EPS_PIXELS evaluates to 3.
C:\projects\opencv\dev\modules\dnn\test\test_model.cpp(260): error: The difference between box.size.width and gtBox.size.width is 22.352630615234375, which exceeds EPS_PIXELS, where
box.size.width evaluates to 294.48736572265625,
gtBox.size.width evaluates to 316.83999633789063, and
EPS_PIXELS evaluates to 3.
[  FAILED  ] Test_Model.TextDetectionByEAST/1, where GetParam() = CUDA/CUDA_FP16 (5398 ms)
force_builders=Custom,Custom Win,Custom Mac
build_image:Custom=ubuntu-openvino-2021.1.0:20.04
build_image:Custom Win=openvino-2021.1.0
build_image:Custom Mac=openvino-2021.1.0

test_modules:Custom=dnn,python2,python3,java
test_modules:Custom Win=dnn,python2,python3,java
test_modules:Custom Mac=dnn,python2,python3,java

buildworker:Custom=linux-1
test_opencl:Custom=ON
test_bigdata:Custom=1
test_filter:Custom=*EAST*

test_opencl:Custom Win=ON
test_filter:Custom Win=*EAST*

@alalek
Copy link
Copy Markdown
Member Author

alalek commented Dec 10, 2020

👍

@opencv-pushbot opencv-pushbot merged commit fce8d8e into opencv:master Dec 10, 2020
@alalek alalek mentioned this pull request Apr 9, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants